INDEX

Explanations

words related to conflicts or potential threats

oai_token-act-pair · gpt-3.5-turbo

instances of the substring "sh" in words

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Ajax

-0.79

OLOGY

-0.75

 Leth

-0.75

ocene

-0.75

ãĥ¼ãĥĨãĤ£

-0.72

DragonMagazine

-0.70

 è£ıè

-0.70

etheless

-0.69

 Reach

-0.68

 Luther

-0.67

POSITIVE LOGITS

atters

1.28

apesh

1.16

attering

1.14

ippers

1.14

ooting

1.13

ackle

1.11

anks

1.11

rapnel

1.09

ipped

1.09

ipping

1.09

Activations Density 0.013%

words related to conflicts or potential threats

instances of the substring "sh" in words

No Comments

No Known Activations